Authorship Weightage Algorithm for Academic publications: A new calculation and ACES webserver for determining expertise.

Abstract

Finding experts in any field can be difficult, especially when relying on academic publications given the use of jargons despite publication lists being publicly available. Within the use of publications, discernment of expertise by authorship positions is often absent in the many pub-lication-based expert search platforms available which could function as another discernment filter of expertise. Given that it is common in many academic fields for the research group lead or lab heads to take the position of the last author, the existing authorship scoring systems that assign a decreasing weightage from the first author would not reflect the last author correctly. To address these mentioned problems, we incorporated natural language processing (Common Crawl using fastText) to identify related keywords for a search compatible to using jargons as well as an au-thorship positional scoring with the option to provide greater weightage to the last author. The resulting output is a ranked scoring system of researchers upon every search which we imple-mented as a webserver for internal agency use called ‘APD lab Capability & Expertise Search (ACES)’ webserver which can be accessed from webserver.apdskeg.com/aces.